arXiv subjects

Biao Qin

Publications and source records attributed to Biao Qin.

9 recordsLinked to original sources

arXiv2026

Writer-R1: Enhancing Generative Writing in LLMs via Memory-augmented Replay Policy Optimization

As a typical open-ended generation task, creative writing lacks verifiable reference answers, which has long constrained reward modeling and automatic evaluation due to high human annotation costs, evaluative bias, and coarse feedback signals. To address these challenges, this paper first designs a multi-agent collaborative workflow based on Grounded Theory, performing dimensional decomposition and hierarchical induction of the problem to dynamically produce interpretable and reusable fine-grained criteria. Furthermore, we propose the Memory-augmented Replay Policy Optimization (MRPO) algorithm: on the one hand, without additional training, MRPO guides models to engage in self-reflection based on dynamic criteria, enabling controlled iterative improvement; on the other hand, we adopt the training paradigm that combines supervised fine-tuning with reinforcement learning to convert evaluation criteria into reward signals, achieving end-to-end optimization. Experimental results demonstrate that the automatically constructed criteria achieve performance gains comparable to human annotations. Writer-R1-4B models trained with this approach outperform baselines across multiple creative writing tasks and surpass some 100B+ parameter open-source models.

cs.CL↗

arXiv2026

QChunker: Learning Question-Aware Text Chunking for Domain RAG via Multi-Agent Debate

The effectiveness upper bound of retrieval-augmented generation (RAG) is fundamentally constrained by the semantic integrity and information granularity of text chunks in its knowledge base. To address these challenges, this paper proposes QChunker, which restructures the RAG paradigm from retrieval-augmentation to understanding-retrieval-augmentation. Firstly, QChunker models the text chunking as a composite task of text segmentation and knowledge completion to ensure the logical coherence and integrity of text chunks. Drawing inspiration from Hal Gregersen's "Questions Are the Answer" theory, we design a multi-agent debate framework comprising four specialized components: a question outline generator, text segmenter, integrity reviewer, and knowledge completer. This framework operates on the principle that questions serve as catalysts for profound insights. Through this pipeline, we successfully construct a high-quality dataset of 45K entries and transfer this capability to small language models. Additionally, to handle long evaluation chains and low efficiency in existing chunking evaluation methods, which overly rely on downstream QA tasks, we introduce a novel direct evaluation metric, ChunkScore. Both theoretical and experimental validations demonstrate that ChunkScore can directly and efficiently discriminate the quality of text chunks. Furthermore, during the text segmentation phase, we utilize document outlines for multi-path sampling to generate multiple candidate chunks and select the optimal solution employing ChunkScore. Extensive experimental results across four heterogeneous domains exhibit that QChunker effectively resolves aforementioned issues by providing RAG with more logically coherent and information-rich text chunks.

cs.CL↗

arXiv2025

Invoke Interfaces Only When Needed: Adaptive Invocation for Large Language Models in Question Answering

The collaborative paradigm of large and small language models (LMs) effectively balances performance and cost, yet its pivotal challenge lies in precisely pinpointing the moment of invocation when hallucinations arise in small LMs. Previous optimization efforts primarily focused on post-processing techniques, which were separate from the reasoning process of LMs, resulting in high computational costs and limited effectiveness. In this paper, we propose a practical invocation evaluation metric called AttenHScore, which calculates the accumulation and propagation of hallucinations during the generation process of small LMs, continuously amplifying potential reasoning errors. By dynamically adjusting the detection threshold, we achieve more accurate real-time invocation of large LMs. Additionally, considering the limited reasoning capacity of small LMs, we leverage uncertainty-aware knowledge reorganization to assist them better capture critical information from different text chunks. Extensive experiments reveal that our AttenHScore outperforms most baselines in enhancing real-time hallucination detection capabilities across multiple QA datasets, especially when addressing complex queries. Moreover, our strategies eliminate the need for additional model training and display flexibility in adapting to various transformer-based LMs.

cs.CL↗

arXiv2025

LEADRE: Multi-Faceted Knowledge Enhanced LLM Empowered Display Advertisement Recommender System

Display advertising provides significant value to advertisers, publishers, and users. Traditional display advertising systems utilize a multi-stage architecture consisting of retrieval, coarse ranking, and final ranking. However, conventional retrieval methods rely on ID-based learning to rank mechanisms and fail to adequately utilize the content information of ads, which hampers their ability to provide diverse recommendation lists. To address this limitation, we propose leveraging the extensive world knowledge of LLMs. However, three key challenges arise when attempting to maximize the effectiveness of LLMs: "How to capture user interests", "How to bridge the knowledge gap between LLMs and advertising system", and "How to efficiently deploy LLMs". To overcome these challenges, we introduce a novel LLM-based framework called LLM Empowered Display ADvertisement REcommender system (LEADRE). LEADRE consists of three core modules: (1) The Intent-Aware Prompt Engineering introduces multi-faceted knowledge and designs intent-aware pairs that fine-tune LLMs to generate ads tailored to users' personal interests. (2) The Advertising-Specific Knowledge Alignment incorporates auxiliary fine-tuning tasks and Direct Preference Optimization (DPO) to align LLMs with ad semantic and business value. (3) The Efficient System Deployment deploys LEADRE in an online environment by integrating both latency-tolerant and latency-sensitive service. Extensive offline experiments demonstrate the effectiveness of LEADRE and validate the contributions of individual modules. Online A/B test shows that LEADRE leads to a 1.57% and 1.17% GMV lift for serviced users on WeChat Channels and Moments separately. LEADRE has been deployed on both platforms, serving tens of billions of requests each day.

cs.IR↗

arXiv2024

CoheDancers: Enhancing Interactive Group Dance Generation through Music-Driven Coherence Decomposition

Dance generation is crucial and challenging, particularly in domains like dance performance and virtual gaming. In the current body of literature, most methodologies focus on Solo Music2Dance. While there are efforts directed towards Group Music2Dance, these often suffer from a lack of coherence, resulting in aesthetically poor dance performances. Thus, we introduce CoheDancers, a novel framework for Music-Driven Interactive Group Dance Generation. CoheDancers aims to enhance group dance generation coherence by decomposing it into three key aspects: synchronization, naturalness, and fluidity. Correspondingly, we develop a Cycle Consistency based Dance Synchronization strategy to foster music-dance correspondences, an Auto-Regressive-based Exposure Bias Correction strategy to enhance the fluidity of the generated dances, and an Adversarial Training Strategy to augment the naturalness of the group dance output. Collectively, these strategies enable CohdeDancers to produce highly coherent group dances with superior quality. Furthermore, to establish better benchmarks for Group Music2Dance, we construct the most diverse and comprehensive open-source dataset to date, I-Dancers, featuring rich dancer interactions, and create comprehensive evaluation metrics. Experimental evaluations on I-Dancers and other extant datasets substantiate that CoheDancers achieves unprecedented state-of-the-art performance. Code will be released.

cs.SD↗

arXiv2024

Meta-Learning Empowered Meta-Face: Personalized Speaking Style Adaptation for Audio-Driven 3D Talking Face Animation

Audio-driven 3D face animation is increasingly vital in live streaming and augmented reality applications. While remarkable progress has been observed, most existing approaches are designed for specific individuals with predefined speaking styles, thus neglecting the adaptability to varied speaking styles. To address this limitation, this paper introduces MetaFace, a novel methodology meticulously crafted for speaking style adaptation. Grounded in the novel concept of meta-learning, MetaFace is composed of several key components: the Robust Meta Initialization Stage (RMIS) for fundamental speaking style adaptation, the Dynamic Relation Mining Neural Process (DRMN) for forging connections between observed and unobserved speaking styles, and the Low-rank Matrix Memory Reduction Approach to enhance the efficiency of model optimization as well as learning style details. Leveraging these novel designs, MetaFace not only significantly outperforms robust existing baselines but also establishes a new state-of-the-art, as substantiated by our experimental results.

cs.GR↗

arXiv2024

Pseudo-Label Calibration Semi-supervised Multi-Modal Entity Alignment

Multi-modal entity alignment (MMEA) aims to identify equivalent entities between two multi-modal knowledge graphs for integration. Unfortunately, prior arts have attempted to improve the interaction and fusion of multi-modal information, which have overlooked the influence of modal-specific noise and the usage of labeled and unlabeled data in semi-supervised settings. In this work, we introduce a Pseudo-label Calibration Multi-modal Entity Alignment (PCMEA) in a semi-supervised way. Specifically, in order to generate holistic entity representations, we first devise various embedding modules and attention mechanisms to extract visual, structural, relational, and attribute features. Different from the prior direct fusion methods, we next propose to exploit mutual information maximization to filter the modal-specific noise and to augment modal-invariant commonality. Then, we combine pseudo-label calibration with momentum-based contrastive learning to make full use of the labeled and unlabeled data, which improves the quality of pseudo-label and pulls aligned entities closer. Finally, extensive experiments on two MMEA datasets demonstrate the effectiveness of our PCMEA, which yields state-of-the-art performance.

cs.LG↗

arXiv2024

Sliding ferroelectric memories and synapses

Ferroelectric materials with switchable electric polarization hold great promise for a plethora of emergent applications, such as post-Moore's law nanoelectronics, beyond-Boltzmann transistors, non-volatile memories, and above-bandgap photovoltaic devices. Recent advances have uncovered an exotic sliding ferroelectric mechanism, which endows to design atomically thin ferroelectrics from non-ferroelectric parent monolayers. Although notable progress has been witnessed in understanding its fundamental properties, functional devices based on sliding ferroelectrics, the key touchstone toward applications, remain elusive. Here, we demonstrate the rewritable, non-volatile memory devices at room-temperature utilizing a two-dimensional (2D) sliding ferroelectric semiconductor of rhombohedral-stacked bilayer molybdenum disulfide. The 2D sliding ferroelectric memories (SFeMs) show superior performances with a large memory window of >8V, a high conductance ratio of above 106, a long retention time of >10 years, and a programming endurance greater than 104 cycles. Remarkably, flexible SFeMs are achieved with state-of-the-art performances competitive to their rigid counterparts and maintain their performances post bending over 103 cycles. Furthermore, synapse-specific Hebbian forms of plasticity and image recognition with a high accuracy of 97.81% are demonstrated based on flexible SFeMs. Our work demonstrates the sliding ferroelectric memories and synaptic plasticity on both rigid and flexible substrates, highlighting the great potential of sliding ferroelectrics for emerging technological applications in brain-inspired in-memory computing, edge intelligence and energy-efficient wearable electronics.

cond-mat.mes-hall↗

arXiv2022

Local Differential Privacy for Belief Functions

In this paper, we propose two new definitions of local differential privacy for belief functions. One is based on Shafer's semantics of randomly coded messages and the other from the perspective of imprecise probabilities. We show that such basic properties as composition and post-processing also hold for our new definitions. Moreover, we provide a hypothesis testing framework for these definitions and study the effect of "don't know" in the trade-off between privacy and utility in discrete distribution estimation.

cs.CR↗